Lessons from the Great Reflation
Monetary policy is under the cosh. Where did it all go wrong?
INFLATION SHOCKS ARE LIKE LONDON BUSES—you wait over a decade for one to come along and three arrive at once.
And so, as of early-2023, Britain is navigating the aftermath of Brexit, a global pandemic, and the near-unprecedented energy shock from Russian irredentism. These shocks have combined to push up headline consumer prices by double digits since the summer and core inflation by above 6 percent more recently—the highest since the aftermath of the ERM crisis in 1992.
Though the shocks are weighed differently, the challenge is similar elsewhere.
The pandemic-related fiscal-monetary shock was greatest in the United States, the energy shock largest for mainland Europe, while the UK sits somewhere between the two—plus Brexit.
The narrative repeats in other countries—or is felt simply through spillovers to global traded goods prices and supply chains. Even Japan is showing signs of inflationary life.
In Australia, where CPI has followed the global trend upwards to reach 7.8%YoY in Q4, the Treasury announced last July a “wide‑ranging review of the Reserve Bank of Australia” to be completed next month. The report provides “an important opportunity to ensure that our monetary policy framework is the best it can be, to make the right calls in the interests of the Australian people and their economy.” A website accompanies this review.
Others will rightly question the origins of our Great Reflation—and the size and speed of the associated policy response.
Abstracting from the energy and Brexit shocks, how might we reflect upon the pandemic-related monetary shock, reflation, and associated monetary policy response? Where did it all go wrong?
In short, as during and after the Great Financial Crisis (GFC), macroeconomics has been found wanting. This time, dropping money from monetary macroeconomics left central banks asleep at the wheel when inflation hit.
We pointed out this problem more than two years ago, in one of the first posts on this blog, with the title “The big myth about money and inflation”
The key point we made at the time (end-2020), was:
If money printing is associated with actual demand generation (via government spending, or transfers), then the link into inflation is likely to be much stronger.
This is a simple difference, between different types of monetary expansion. But somehow it has been missed by many, leading to ‘forecasting disasters’ by those who are ‘religious’ about the role of money in inflation dynamics.”
We wish we had made the argument more forcefully, and repeated it more often, as it turned out to be rather crucial for global macro trends in 2021-2022.
What happened to money?
The pandemic-related expansion in broad—and base—money has surely played a crucial, though not the exclusive, role in the recent acceleration in inflation. Only last week the Bank for International Settlements concluded such a link exists.
There is indeed evidence that money growth and inflation have been closely linked recently. For example, across countries, there is a statistically and economically significant positive correlation between excess money growth in 2020 and average inflation in 2021 and 2022.
It is natural therefore to question how money features in the calibration of monetary policy today—during and since the pandemic.
Unfortunately, there is little calibration to such money aggregates in today’s state-of-the-art in macro models. Why?
Well, exactly 20 years ago the publication of Michael Woodford’s Interest and Prices: Foundations of a Theory of Monetary Policy sought “to provide theoretical foundations for a ruled-based approach to monetary policy.” This book, and associated research, had an enormous impact on academic macroeconomics and central banking.
Partly this agenda was only research aping central banking practice. But Interest and Prices and associated neo-Keynesian research legitimised the view that money can be largely ignored in macroeconomic modelling and policymaking.
Consider the related paper presented at the ECB in 2006, on the role of money in the conduct of monetary policy, later published in the JMCB, where Woodford quotes former-Fed governor Laurence Meyer insisting that “money plays no explicit role in today’s consensus macro model, and it plays virtually no role in the conduct of monetary policy.”
This quote was lifted from a 2002 speech by then-Bank of England Deputy Governor Mervyn King on the role of money in the economy—more on which later.
Crucially, in this 2006 paper—“How Important is Money in the Conduct of Monetary Policy?”—Woodford questioned the ECB’s two-pillar approach to monetary policy, in particular the separation of economic analysis from monetary developments in forecasting and policy.
This is because the research agenda in Interest and Prices did not require money to be explicitly modelled—and money was ignored in a cashless limit.
The paradigm combines sticky prices and dynamic stochastic general equilibrium (DSGE) modelling—allowing for the “endogeneity of private-sector expectations”—with simple monetary policy rules to analyse inflation control and stabilisation policy.
But why no money? Well, the simplest answer—and the one given in Chapter 1 of Woodford’s book—is that the practice of monetary policy was, and is still, concerned with the choice of interest rate to achieve policy objectives (e.g., an inflation target.) The policy rate is chosen exogenously and (base) money adjusts endogenously, through open-market operations, to achieve this target rate.
Thus was born “Keynesian macroeconomics without the LM curve.”
The theoretical corollary of dropping money was the observation that a money demand function can always be grafted onto the model—with no impact on the resulting solution for the price level.
The textbook treatment by Jordi Galí of UPF and Barcelona School of Economics, explains that when added, money demand plays “no independent role in determining the equilibrium;” instead money adjusts endogenously given real output, the price level, and interest rate (p. 28).
Crucially, the framework assumes that money demand always equals supply.
Yet occasions have emerged historically when an imbalance might emerge between the two, something contemplated seriously in the 1970s and worth exploring again.
What if money supply does not at all times equal money demand?
This seems like an awkward question today, as we usually think markets continuously clear. But such disequilibrium was considered a serious possibility during the Great Inflation in the 1970s; money might be a buffer for the adjustment of the community to structural changes in the financial system.
In “Disequilibrium money — a note” Charles Goodhart reviewed this possibility while at the Bank of England in the early 1980s—later made public as part of a collection of essays.
As Goodhart explains, disequilibrium money was popularised by David Laidler and others during the breakdown of money demand functions in the 1970s:
The essence of this approach is to regard money holdings as in some large part being used as a buffer stock… economic agents may rationally and optimal decide to respond to the continuing stream of developments… not by a thoroughgoing, continuous reconsideration of their full economic dispositions, but allowing such shocks to impinge initially upon certain assets/liabilities whose characteristics make them suitable to act as buffers.
And money holdings are
pre-eminently suitable to be used by agents as buffer stocks… shocks in all other markets will tend to have a counterpart effect on cash flows. For example, if goods are rationed and subject to price control, as occurred during wartime and in socialist economies, the initial effect of that is likely to find a reflection in holdings of money.
Goodhart did not imagine the impact of a pandemic and related lockdowns. But a period of rationing provides a decent enough metaphor, when government fiscal-monetary support was the counterpart to a build up of private saving—taken largely in monetary form thanks to central bank balance sheet expansion.
Importantly, if money indeed serves a a buffer in times of upheaval, supply need not equal money demand—creating disequilibrium at the heart of household balance sheets.
This issue was raised again in 2007 when Goodhart—in a lecture in honour of Maurice Peston, published by the National Institute of Economic and Social Research—took on directly Woodford’s argument at the ECB.
In this lecture—“Whatever became of the Monetary Aggregates?”; free here—Goodhart warns that the “downgrading of the role of the monetary aggregates in current models, and in forecasting future inflation, has gone too far.”
Why so? First recall the problem with the New Keynesian DSGE paradigm:
There appears to be no need to look at what is happening to money in [Woodford’s] system to achieve the main macro-economic variables of importance to social welfare.
In practice, … a demand-for-money function is fully consistent with the above three equations… [but with such a demand shock] … you learn nothing more from looking at money than you already knew from looking at inflation, output and interest rates.
Goodhart goes on:
But it is not true that all shocks to money are demand-side. The bulk of money is in the form of commercial bank liabilities, and banks can behave very differently over time… If there is a supply shock to money, with certain groups now getting more, or less, access to funding, for example when banks provide mortgages to a wider group of households on easier terms, will this not feed back into the IS curve? Of course it will.
Goodhart is not thinking of goods rationing here, or pandemic lockdowns, but money from the changing disposition of banks in credit provision—an echo of the structural changes to the United Kingdom monetary system in the 1970s. These changes (“regulatory and structural”) brought about “consequential and autonomous changes in bank behaviour” coincident with the breakdown of money demand relations at that time, according to his disequilibrium essay.
We lament parenthetically that in 2007, in his response to Woodford, Goodhart does not reference this disequilibrium paper.
In any case, Goodhart went on to worry about overlooking the impact of money supply changes on wealth, through asset prices such as housing, and the long-run link between such asset prices and expenditure—more on which shortly.
Goodhart ends the paper with a decision tree for thinking about money shocks.
If money growth is not consistent with the paths for output, inflation and interest rates, then the policymaker should ask if it is a money demand or supply shock. If the former, it can be ignored. If the latter, it matters—and ought to elicit a policy response.
In short, as Goodhart warned, state-of-the-art monetary models of two decades ago—those that infiltrated the temples of central banking—could leave policymakers blindsided in the event of a shock to the money supply—especially if money serves as a buffer for saving, exactly as occurred during the pandemic.
But if we allow for such a buffer, how should it be managed?
Unwinding monetary overhangs
There is a large literature on managing “monetary overhangs” such as those that followed the great wars of the last century or the post-communist transition. These overhangs found expression in shortages of goods and incipient inflation as households attempted to cash in cumulated savings.
The analogy to the post-pandemic challenge is striking. Yet the absence of policy discussion of our monetary disequilibrium, drawing on prior experience, is equally striking. Perhaps the shock and overhang wasn’t quite as large. But the disequilibrium was considerable still.
Consider the discussion of quantitative tightening in recent years. This was largely conducted in terms of fine tuning the policy stance or, with the Bank of England, to making space for future upward expansion. It has not been framed in terms of calibrating policy to the unwind of the household monetary disequilibrium.
It’s worth briefly recalling how such overhangs were thought to be resolved.
Indeed, for the late Rudiger Dornbusch, such overhangs could be managed in two ways. First, a “big, possibly one-shot inflation” could be used to return real money balances to their equilibrium levels. Second, the liquid assets created previously could be “immobilised” by being “written off, consolidated as debt or retired by asset exchanges or budget surpluses.”
Curiously, according to Dornbusch, once this stock problem is resolved, inflation will only continue if there is an underlying flow problem that remains. And it has been found that historically, “the release of a monetary overhang does not result in permanent inflation, much less a hyperinflation.”
Seen in this light, the monetary policy debate fell short.
There was a monetary disequilibrium. This could be dealt with through a combination of temporary, though perhaps multi-year, inflation alongside quantitative tightening to withdraw the build up of money and reshape private portfolios. But there is no need to panic about persistent inflation unless a residual flow imbalance remained once the pandemic was resolved.
But this discussion could not happen, of course, because central banks dropped money aggregates two decades before. Instead of being proactive, Inflation forecast targeting central banks became reactive in setting policy to incoming inflation—despite the fact this inflation was baked in the cake due to the pandemic intervention.
Once we acknowledge a monetary disequilibrium exists, it needs to be unwound. But by what mechanism does it transmit? This is where the monetarists come into their own.
For all that Milton Friedman wrote about money, it is difficult to find a clear distinction between the source of the money supply shock and the resulting adjustment. For example, should we should distinguish for analytical purposes between a shock to high powered—or outside—as opposed to broad money? Recent experience suggests we should.
In his Theoretical Framework for Monetary Analysis in 1970, Friedman devotes only about half a page of the forty-five in total to the drivers of the money supply. For Friedman, at that stage at least, the quantity theory of money was sacrosanct—further details could be ignored.
Still, taking as given the money supply shock, the monetarists can be said to provide the richest description of the resulting transmission throughout the macroeconomy.
Consider first the traditional Keynesian description of monetary transmission. Again, here’s Goodhart (from a different paper, page 161):
The effect of a change in the money supply [due to open market operations in the Keynesian framework] is seen to be like a ripple passing along the range of financial assets, diminishing in amplitude and in predictability as it proceeds farther along from the initial disturbance. This ‘ripple’ eventually reaches to the long end of the financial market, causing a change in yields, which will bring about a divergence between the cost of capital and the return on capital.
Hence money transmits through the yield curve into investment decisions, then investment and activity—and, perhaps, inflation.
For the monetarists there was a much richer process in play.
Drawing on detailed studies with Anna Schwartz, Milton Friedman’s description in his Wincott Lecture in 1970 is as clear as anything we can hope to imagine:
One of the most difficult things to explain in simple fashion is the way in which a change in the quantity of money affects income. Generally, the initial effect is not on income at all, but on the prices of existing assets, bonds, equities, houses, and other physical capital. This effect, the liquidity effect stressed by Keynes, is an effect on the balance-sheet, not on the income account. An increased rate of monetary growth, whether produced through open-market operations or in other ways, raises the amount of cash that people and businesses have relative to other assets. The holders of the now excess cash will try to adjust their portfolios by buying other assets. But one man’s spending is another man’s receipts. All the people together cannot change the amount of cash all hold—only the monetary authorities can do that.
He continues, original emphasis:
However, as people attempt to change their cash balances, the effect spreads from one asset to the another. This tends to raise the prices of assets and to reduce interest rates, which encourages spending to produce new assets and also encourages spending on current services rather than on purchasing existing assets. That is how the initial effect on balance-sheets gets translated into an effect on income and spending.
In summary, Friedman noted:
...monetarists insist that a far wider range of assets and of interest rates must be taken into account. They give importance to such assets as durable and even semi-durable consumer goods, structures and other real property.
Or, returning to Goodhart’s survey, Keynesians:
expect people to buy financial assets when they feel that they have larger money balances than they strictly require (given the pattern, present or prospective, of interest rates), whereas monetarists expect the adjustment to take place through 'direct' purchases of a wider range of assets, including physical assets such as consumer durables. (Page 164.)
For monetarists, then, a money shock feeds slowly into income and prices, first into physical and financial assets, durable and non-durable goods, and through balance sheets to income and spending. This happens echo Goodhart’s conclusions in response to Woodford’s cashless approach, though he was careful to insist at the time: “I am far from being a card-carrying monetarist.”
If this description best fits the transmission of a monetary shock to income, can we visualise it?
The exceptional increase in household wealth during the pandemic was noted here previously (here and here.) Household net worth expanded by USD34 trillion from end-2019 through end-2021, an increase of >150% of end-2019 GDP in only 2 years.
This wealth increase was mainly due to the increase in financial and non-financial assets (housing).
The chart below shows the change in the value of outstanding assets held by households for select items from end-2019. Currency and deposits held by households began the sharp increase in the first quarter of 2020, during the initial lockdowns, and increased by about USD4½ trillion in total. The increase in equity and real estate assets, mainly due to valuation adjustments, followed—and in value terms was a much larger influence on household net worth.
At its peak, equity and investment fund shares, which fell in value by USD6 trillion by end-March 2020, increased nearly USD18 trillion by end-2021 on end-2019, or USD24 trillion peak-to-trough. The declines experienced last year during the interest rate normalisation have weighed on such wealth since, but the levels remained above pre-pandemic norms.
Real estate assets continued to increase in value still in 2022Q3, though at a diminishing pace as housing feels the strain of higher rates. Still, nearly USD13 trillion was added to household wealth through real estate since end-2019.
Put another way, at its peak, every additional USD1 trillion in monetary wealth during the pandemic came to be reflected in roughly USD6 trillion in non-monetary wealth, mainly equities and housing.
Alternatively, we can visualise how the increase in household nominal money balances fed various prices—as below, note the different scales. Within the Personal Consumption Expenditure (PCE) deflator, durable goods accelerated first, though lagging equities (seen above), followed by non-durable goods and, lastly, services. Average hourly earnings jumped initially on compositional effects, but has since kept up with the PCE deflator.
Our Great Reflation has taken many by surprise. Here we have reflected on the monetary aspects—downplaying the energy shock and Brexit challenge.
Central banks in particular have been found reactive, responding to incoming inflation rather than inflation targeting proper. That’s a sign a panic.
One reason for this panic is the new Keynesian work program that reached some kind of peak twenty years ago with Woodford’s Interest and Prices. This approach encouraged the idea that money could be ignored—but as Goodhart warned at the time, there was a risk that the implications of a money supply shock would be missed. And the pandemic created for the community a huge money supply shock—and this disequilibrium is being unconsciously unwound through inflation.
It will be interesting to read the official post mortems for our recent experience.
The first chance to reflect comes in Australia next month, with the review of the RBA. Skimming through the main macro model used by the RBA, money doesn’t appear as an input for forecasting—though perhaps it is hidden somewhere in the annex. They too have been a victim of the new Keynesian group think.
We might also note some reflection already in the UK last summer, when the Bank’s Chief Economist Huw Pill asked: “What did the monetarists ever do for us?”
Pill takes a step towards revising the role of money in these models:
The analysis I have presented explores whether monetary quantities play an additional role in the transmission of monetary policy to aggregate spending (and thus pricing) decisions, beyond that already captured by the impact of changes in the official policy interest rate.
Expressed in the context of the canonical new Keynesian model of monetary policy transmission, this essentially revolves around whether it is sufficient for the policy rate alone to appear in the ‘dynamic IS curve’ equation in that model, or whether other measures of monetary and financial conditions are required.
In short, Pill suggests perhaps the dynamic IS curve might need updating. It’s a pity he does not go further. Recall that Goodhart wondered how a money supply shock might “feed back into the IS curve” through asset prices and spending.
In other words, the household balance sheet matters—not just money, but also due to the knock on of money supply shocks—and this should be reflected in our state-of-the-art thinking.
Curiously, Pill’s speech draws heavily on the 2002 speech by Mervyn King mentioned above and referenced by Woodford.
In reflecting upon our recent monetary malaise, therefore, we might recall once more the fact that monetary macroeconomics continually fails to integrate balance sheets and financial structures. This time, central banks have been blindsided in the pandemic as household balance sheets strengthened—a decade ago, fiscal policy was calibrated without due regard for household balance sheet weakness.
Finally, we end on an even more unhappy note: none of the above can be used to endorse the policy tightening response over the past year—which has been based on panic in the absence of a structured analytical framework. And so the prospect of further policy errors abound.
The content in this piece is partly based on proprietary analysis that Exante Data does for institutional clients as part of its full macro strategy and flow analytics services. The content offered here differs significantly from Exante Data’s full service and is less technical as it aims to provide a more medium-term policy relevant perspective. The opinions and analytics expressed in this piece are those of the author alone and may not be those of Exante Data Inc. or Exante Advisors LLC. The content of this piece and the opinions expressed herein are independent of any work Exante Data Inc. or Exante Advisors LLC does and communicates to its clients.
Exante Advisors, LLC & Exante Data, Inc. Disclaimer
Exante Data delivers proprietary data and innovative analytics to investors globally. The vision of exante data is to improve markets strategy via new technologies. We provide reasoned answers to the most difficult markets questions, before the consensus.
This communication is provided for your informational purposes only. In making any investment decision, you must rely on your own examination of the securities and the terms of the offering. The contents of this communication does not constitute legal, tax, investment or other advice, or a recommendation to purchase or sell any particular security. Exante Advisors, LLC, Exante Data, Inc. and their affiliates (together, "Exante") do not warrant that information provided herein is correct, accurate, timely, error-free, or otherwise reliable. EXANTE HEREBY DISCLAIMS ANY WARRANTIES, EXPRESS OR IMPLIED.
Simplifying assumptions- as there is “one” interest rate that mattets- must live or die or their ability to predict. And obviously the new k paradigm has suffered its second obvious failure- after 2008.
Really good. I am old enough to have missed in my formative years the new keynesian group think. We were thought to think of policy as affecting various “channels of transmission”, and besides knowing based on theory what the first derivatives should be, the rest was more descriptive.